https://www.youtube.com/watch?v=N5YDkQ6oFcE

https://www.heraldextra.com/news/2026/a ... tage-days/

I had a long post, but it keeps getting eaten up by server delays, so.......... more later.

I am not sure if that youtube creator channel is real or fake?

Look,
it is 16 year old channel,
14 years with mostly not narrated railfan videos,
after that videos of obscure topics with maybe AI generated voice...
plus asking for patreon money on the end of the videos?
For me that is a hacked account without any proof of the working.
Clickbait and money the only goal!

Is here some real people who was in that museum lately?

It’s definitely a real channel by a real person.

Great Basin Car Shops channel is ran by a good friend of mine, and I've even helped him film some of the videos with him riding shotgun in my car sticking a camera out the windows chasing Rio Grande Southern 20 in Durango (curiously he was accused of AI on that too, which had us both howling the rest of the trip freezing in snow waiting for RGS 20 to pass and we'd crack jokes like "this AI is SO convincing, my legs really can't feel anything!") He rebranded the channel about two years ago which is why the content style shifted, but trust me it's very human made. I saw photos of the Lehi museum trip before the video was made in a group chat and we were all a bit flabbergasted by them.

There is a group in Ohio right now trying to raise money for locomotive repairs. The picture in their FB post and on their website is an AI-generated shot of the locomotive flying around a curve with no one in the cab, no headlight and sparks flying off the wheels. Whoever posted it says it was meant to be "fun". Most of the replies are negative but the attitude of the group's responses is "we don't care what you think about it, just send us money!" The leader of the group is perfectly fine with it as well.

Well, I guess that's one way to decide what projects are worth supporting... It's a shame the project will suffer though.

This whole thing is so obnoxious.

AI is a really good tool when used responsibly. It's how I built the Catalogr tool that lets the CRHS distribute cataloging duties along with a whole bunch of digital infrastructure and tools for other groups. The New Haven Archive (https://archive.nhrhta.org/), for example, wouldn't be possible if it were for GenAI (specifically Vercel's V0.app IDE).

But, and this is the important part, you'd NEVER know. That's because the tools were used responsibly and intelligently.

People are freaking out aboug GenAI, in some cases legitimately, but we also need to remember that ALL big technological advances bring dangers. Hell, the Best Friend of Charleston killed someone in its own first six months of operation.

In that instance, like this one, the operators of the technology did something ill advised with it. In one case it was tying down a safety valve while the other was making and passing slop off as historical fact.

I think the REAL threat from GenAI is that the harmful repercussions from its poor use are far less obvious than a boiler explosion, although possibly more harmful to more people.

https://www.facebook.com/tlewrm/posts/p ... L8W4wRp7zl I saw that, and I was positively astounded by the over-the-top reaction to something about as bad as photoshopping a smiley face on a unit a la Thomas. Definitely a mob dynamic developing, where if you don't agree that AI is evil you are morally suspect. I am reminded of the scene in Voltaire's Candide where the main character, visiting a protestant area of Europe, is asked if he believes that the pope is antichrist, and when he isn't sure that the pope is antichrist, and hadn't even thought about it, the people are outraged, dump chamber pots on his head, etc., and essentially run him out of town.

As both a graphic artist and musician, I'm pretty much vehemently against AI in the creative arts. Do note that I think AI does have its uses. Just not in these cases.

However, in my opinion, the TLE&W post isn't as egregious as some make it out to be. Would I have used AI in that post? Big nope. But I'm not hideously offended as it's not grossly portraying something it's not. Is it a "fun" graphic? Maybe so, maybe not. Though, again in my opinion, it just doesn't fit the campaign.

As far as the Lehi museum goes, if true, extensive use of generative AI to make historic displays is such a bad move - kinda like making up history. AI does no research. It extrapolates and regurgitates what is in its training data - none of which has been screened in any way to achieve accuracy in the final output.

I'm also a big fan of requiring clear labelling of AI content.

I know the creator, and will vouch for him... He is a very good researcher, an amateur archeologist, very knowledgeable on southwest and Utah history, as well as early Utah/Colorado railroad history.

AI may in the future make contributions to society... but for now its mostly a plaything. I have seen very good graphics produced by AI... by careful artists who rework and rework to get it right. I have also seen a significant number of badly restored and colorized historic photos, which proport to be accurate, and in the process create false history. In the case of the Lehi Museum, in the video, the museum was primarily about local pioneer families. The museum had closed during covid, and the artifacts placed in storage or dispersed, and in the absence of artifacts the community behind he museum chose (bad) AI to tell local history, much of it based on "old" stories which were highly fictionalized. The result was a comic book masquerading as a museum.

I am a little more sympathetic to museum promotion which uses (frequently bad) AI. A poster is not history, but a bad poster, be it AI or otherwise hand created it a bad poster, and will suggest a sloppy presentation.

Okay. The narrator brings up a lot of valid points. Not only is some of the A.I. imagery used in the museum blatantly wrong or deceptive, but just using that style of artwork instantly screams "no imagination, no effort, no talent, no knowledge, just type some words and see what happens." I swear, there will come a day when we look back at the short-term prevalence of this "graphic design" with horror and disgust, much as some look back at certain past fashion trends with embarrassment.

BUT:
The blowback from this "viral" video has been horrible.
There are people calling for this place to be defunded, shut down, or to refund any government money involved. Which shows you 1) how many people think museum projects are all government-funded wards of the state, and 2) what a high standard the "online mobs" demand.

I don't know the specifics of this particular place. It's so small that I never heard of it, and I've been tracking rail museums and preservation projects across North America for decades. It's new-ish. So by default I'll cast them SOME benefit of the doubt.

I have seen, or been part of, several groups in the past year or two (few of them railroad-related) where, all of a sudden, some well-meaning volunteer or two, often retired or a student, suddenly "discovers" A.I. processing, and then turn around and produce a social media post or three a day using it, then crank out new brochures, posters, advertisements, and the like using that now-ubiquitous A.I. graphic style that now makes me throw up in my mouth a bit every time I see it. Okay, so yeah, they have more "content" for people to engage with online, and stuff may be getting done that might have cost more money/time had they waited for a real graphic designer or had to hunt down a talented volunteer. But it all starts to look the same between a diverse number of groups and interests; it LOOKS lazy and cheap; and it slowly turns off people with a revulsion to A.I. or the so-called "style" that all comes out of it.

I'm 99% positive this is exactly what happened here. Most likely, ONE overzealous volunteer with a stack of printer paper and frames decided to fill up all the blank space throughout the building. And no one was in a position to proofread or critique the projects for accuracy or implausibility, or call said volunteer(s) "on the carpet." If you know the typical local historical society, the average age of the active members is between 70 and death, and they may be lucky to have anyone at all the least bit computer-literate and social-media-savvy.

Those exhibits can be replaced. Are you offering to replace them? With what? You have better?

The criticism and online blowback as a result of this video has been harsh. I can see that the two entities involved have had to remove some posts to their Facebook pages that got heavily hit by the online lynch mobs instigated by this video.
But the reality is that we are letting the perfect be the enemy of the "good enough." I'd rather see a diverse and happy crowd looking through a restored depot with a bunch of half-a**ed A.I. pictures mixed in among the real things than yet ANOTHER abandoned rail station being left to vandals and ruin and eventual arson or demolition.

Is EVERY ONE Of your rolling stock exhibits perfectly painted and rust-free? Do you dare to have a Rock Island or NYC or C&NW lantern in your ex-CB&Q depot? Why is there a GG1 in Elkhart, or Texas? Are there railroad books from Waldenbooks or Borders or Books-a-Million bargain shelves in your "library"? Did that desk, telephone, and telegraph set really come from that station? Is that model railroad accurately representing your local line, or is it just a loop of HO-scale track?

"Let he who is without sin cast the first stone......."

UPDATE: Ironically, as I was typing the above, the news crossed my feed about Lonnie Bunch, the Secretary (read: head director) of the Smithsonian Institution, resigning or "choosing to retire" (well, he IS 73 years old) amidst the "butting of heads" between the Smithsonian's administrations and the Trump Administration's harsh criticisms of the Smithsonian's "woke" and "politically correct" interpretive direction of the past two decades or so, primarily at the National Museum of American History of late...........

https://www.cnn.com/2026/09/08/politics ... mithsonian

I'm working on getting CEU credit established for selected Cumbres & Toltec Friends work sessions. The CEU coordinator at Adams State in Alamosa recommended I include a 'reflection' exercise in the syllabus; I spent two decades in and around university education, had never heard of such. So, I posed the question to Gemini in the Google search site, and got good information and example grading rubrics that fit the work session situation. Cool beans.

I've attempted to use some of the AI agents for model/railroading and software coding tasks, with mixed results. The most 'interesting' one was submitting a vintage image of the Antonito depot for highlighting of the stone wall mortar lines, so I could accurately model the patterns. After a few iterations (it first wanted to turn the stones to bricks), I got what looked like a decent pattern based on other photographs. But, it replaced the smudges that were peoples' faces in the original with faces from other photographs, including a bearded face for a kid. Yeah, right.

So, my experience tells me that small, specific tasks can work well, larger more ambitious ones, not so much. Proper attribution is a must, if one uses such it should be disclosed. Doing a google search these days puts the Gemini assertions up-front, and those usually have their external references as links, so it really helps as a sifter of search results.

Dang thing is just another tool, albeit one with sharp edges...

More nonsense regarding railroad history in believable bad communication of inaccurate info
(logging railroad/shay related)
https://www.facebook.com/reel/1091075406617596

All this brings up a question: Ain't gonna lie, AI does have some good tools. How many of our organizations have or are coming up with an AI policy to govern its use within the organization. I'm currently developing one to present to our BoD for approval.

Randy, I just saw some photos from a visitor to your former place of employment's open house for their new visitor's center... and there is some pretty awful news regarding the Boulder City's museum's choice of "abominable intelligence" to decorate the space. It's one thing to see it at a small museum like the Lehi one, but that it's reached an NSRM facility stopped me in my tracks. A shame too, they clearly were emulating the LA&SL style station murals you'd see in real preserved stations, but now everything looks awful with that AI sheen.

Not from a museum or historical society, but I caught two railroad groups who SHOULD have known better sharing this example of AI garbage on their Facebook accounts:

https://www.facebook.com/reel/1065757279661962

Both cases I'll chalk up to "undertrained" Facebook page moderators.